Structural Basis of Histone H4 Recognition by P55

Downloaded from genesdev.cshlp.org on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

RESEARCH COMMUNICATION

Structural basis of histone H4 matin remodeling complexes—Nucleosome Remodeling Factor (NURF), and Nucleosome Remodeling and Deacety- recognition by p55 lase (NuRD) (Smith and Stillman 1989; Tyler et al. 1996; Verreault et al. 1996; Martinez-Balbas et al. 1998; Wade et Ji-Joon Song, Joseph D. Garlick, and al. 1998; Zhang et al. 1998). p55 is a component of acetyl- Robert E. Kingston1 transferase (Hat1), histone deacetylase (HDAC1), and histone methyltransferase (Polycomb-Repressive Complex2, Department of Molecular Biology, Massachusetts General PRC2) complexes, and it has been shown that p55 is criti- Hospital, Boston, Massachusetts 02114, USA; Department cal for the function of these complexes (Parthun et al. 1996; of Genetics, Harvard Medical School, Taunton et al. 1996; Hassig et al. 1997; Zhang et al. 1997; Boston, Massachusetts 02114, USA Verreault et al. 1998; Czermin et al. 2002; Kuzmichev et al. 2002; Muller et al. 2002). Furthermore p55 copurifies p55 is a common component of many chromatin-modi- with additional complexes involved in gene regulation fying complexes and has been shown to bind to histones. (Zhang et al. 1997; Korenjak et al. 2004). Here, we present a crystal structure of Drosophila p55 Depleting p55 causes a variety of epigenetic defects bound to a histone H4 peptide. p55, a predicted WD40 (Lu and Horvitz 1998; Hayashi et al. 2004; Taylor-Hard- repeat protein, recognizes the first helix of histone H4 ing et al. 2004; Guitton and Berger 2005). via a binding pocket located on the side of a ␤-propeller In Caenorhabditis elegans, LIN53, the ortholog of p55, structure. The pocket cannot accommodate the histone shows a multivulva (Muv) phenotype resulting from mis- fold of H4, which must be altered to allow p55 binding. expression of vulval cell fates (Lu and Horvitz 1998). The Reconstitution experiments show that the binding pocket Arabidopsis ortholog, MIS1, has been extensively characterized, and loss of function mutants cause defects in is important to the function of p55-containing com- ovule development and dysregulation of flowering time plexes. These data demonstrate that WD40 repeat proteins (Hennig et al. 2003; Kohler et al. 2003; Bouveret et al. use various surfaces to direct the modification of histones. 2006). Although the developmental role of p55 in verte- Supplemental material is available at http://www.genesdev.org. brates and insects is not clear, high expression of RbAp48/46 in the ovary and testis might reflect a func- Received January 18, 2008; revised version accepted March tion in mammalian gametogenesis (Qian and Lee 1995). 13, 2008. Since p55 is required for the function of several essential complexes, it is difficult to define the action of p55 in vertebrates and insects using standard genetics. How- In eukaryotic cells, DNA is hierarchically packaged into ever, definition of the precise structural interactions of higher order structures called chromatin. The basic unit p55 might eventually lead to the design of specific mu- of chromatin is the nucleosome, which is formed from tations that disrupt distinct functions. 146 base pairs of DNA wrapped around a histone oc- p55 contains WD40 repeats that are predicted to form tamer. Chromatin is a dynamic structure and can be modi- a ␤-propeller structure (Ach et al. 1997). The WD40 re- fied in several ways (Li et al. 2007). First, chromatin is peat is a well-known protein–protein interaction do- actively assembled and disassembled by histone chaper- main, and WD40-containing proteins are involved in sig- one complexes. These assembly and disassembly pro- nal transduction, protein degradation, and gene regulation. cesses are tightly coupled with DNA replication and Previous in vitro pull-down experiments demonstrated gene expression (Groth et al. 2007). Second, chromatin that in the context of Hat1, p55 interacts with the first can be covalently modified; N- or C-terminal tails of his- helix of histone H4, which is buried in the canonical tones can be methylated, acetylated, phosphorylated, ad- nucleosome structure (Verreault et al. 1998). This is in enylated, or ubiquitylated. These modifications can be uti- contrast to the way in which another WD40 repeat pro- lized as marks for recruiting effector proteins and might tein,WDR5, binds to histone H3. In that complex, it is also directly alter chromatin folding. Third, chromatin the exposed tail of histone H3 that is recognized by the structure is altered by ATP-dependent chromatin remod- surface of the ␤-propeller structure of WDR5 (Wysocka eling complexes, which can alter DNA accessibility by et al. 2005; Couture et al. 2006; Han et al. 2006; Ruthen- disrupting DNA–histone contacts. burg et al. 2006; Schuetz et al. 2006). p55 (p55 or Nurf55 in fly, RbAp48/46 in human, and p55 has remarkable sequence conservation and is in- MSI1 in plants) is highly conserved from plants to hu- volved in diverse epigenetic processes, suggesting that it man. RbAp48/46 was initially identified as a retinoblas- might serve an essential biological role in each complex toma-associated protein (Huang et al. 1991; Qian et al. in which it is found. However, the molecular and struc- 1993). Subsequent studies showed that p55 is a common tural basis of histone recognition by p55 remains unde- component of many different chromatin-modifying com- fined. Moreover, it is not known what functional role(s) plexes with a variety of functions. p55 is the smallest p55 plays in p55-containing complexes or how one pro- subunit in the Chromatin Assembly Factor 1 (CAF1) com- tein can be integral in so many distinct complexes. Here plex as well as a component of the ATP-dependent chro- we present the crystal structures of free p55 and of p55 bound to a histone H4 peptide. These structures reveal [Keywords: p55; Nurf55; RbAp48; histone; chromatin; WD40] that p55 recognizes histone H4 via a binding pocket lo- 1 Corresponding author. cated on the side of a ␤-propeller structure. p55 appears E-MAIL [email protected]; FAX (617) 643-2119. Article published online ahead of print. Article and publication date are to be preorganized for histone binding, as the free and his- online at http://www.genesdev.org/cgi/doi/10.1101/gad.1653308. tone H4-bound structures are superimposable. However,

GENES & DEVELOPMENT 22:000–000 © 2008 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/08; www.genesdev.org 1 Downloaded from genesdev.cshlp.org on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Song et al. the structure shows that the histone fold of H4 has to be tunately, the 27-amino-acid-long connecting loop in the altered substantially upon p55 binding. In addition, the second blade is disordered in the structure. However, the H4-binding pocket is critical for the activity of p55-con- loop in the seventh blade is composed of 16 amino acids, taining complexes and may be used in different ways in and is well ordered. This loop protrudes from the side of different complexes. These and other data suggest that the ␤-propeller structure and forms one side of a pocket, the WD40 repeat domain can utilize various surfaces for with the other side formed from the N-terminal ␣-helix. recognizing histones and that p55 may serve as a multi- Thus, in addition to the predicted canonical ␤-propeller functional protein interaction platform for the complexes. structure, p55 contains two unusual features: the N-terminal helix and the long connecting loops. Results and Discussion Crystal structure of p55 bound to histone H4 peptide Crystal structure of p55 and histone recognition The crystal structure of isolated Drosophila p55 was deter- Previous work demonstrated that p55 binds to histone mined to 2.9 Å using Single Anomalous Dispersion (SAD) H4 (Verreault et al. 1998). To gain insight into how p55 (Supplemental Table 1). There was one molecule per asym- recognizes histone H4, p55 was cocrystallized with a his- metric unit with a relatively high solvent content (81%). tone H4 peptide. The p55–H4 complex structure The crystal lattice was stabilized by protein–metal 15–41 15–41 was determined to 3.2 Å by molecular replacement using (Cd2+) interactions. The structure of p55 encompasses the p55 structure as the search model (Supplemental Table seven WD40 repeats forming a ␤-propeller structure with 1). Of the 27 amino acids in the histone peptide, only 11 an additional ␣-helix at the N terminus (Fig. 1A). The amino acids (Lys31 to Gly41) are ordered in the structure ␤-propeller structure of p55 is similar to other known (Fig. 1B). This region corresponds to the first helix of the ␤-propeller structures. In fact, the presence of an N-ter- histone fold of histone H4 and agrees well with previous minal ␣-helix is highly reminiscent of the ␤ subunit of binding studies (Verreault et al. 1998). The histone H4 the G protein (Sondek et al. 1996). The N-terminal helix peptide is nestled into a “binding pocket” formed by the of p55 is situated between the first and seventh blades of N-terminal ␣-helix and the “binding loop” emerging the ␤-propeller structure, spans ∼40 Å, and is nearly par- from the seventh blade. The structure of p55 bound to allel to the sevenfold symmetric axis of the WD40 repeats. histone H4 is almost identical to that of free p55, and the Interestingly, a C-terminal one turn-helix is positioned binding loop is well ordered in both structures, suggest- on top of the N-terminal ␣-helix, forming an extended ing that the binding pocket is preformed. helix. p55 appears to use a unique mechanism to recognize An interesting feature of the p55 structure is the pres- and bind histone H4. In this structure of the p55–H4 ence of long connecting loops, which extend from the 15–41 complex, the histone H4 peptide is bound in the binding ␤-strands and protrude far from the core before connect- pocket on the side of the ␤-propeller structure of p55. ing back to the ␤-propeller structure. The loops in the This pocket is entirely composed of regions of the pro- second and seventh blades are particularly long. Unfor- tein that are specific to p55, the N-terminal ␣-helix and the loop extruding from the seventh blade of the ␤-propeller. These additional features may begin to explain the difference in binding modes between p55 and another known histone-binding ␤-propeller protein, WDR5. Both surfaces of the binding pocket are absent in WDR5. In WDR5, histone H3 peptides are bound on the top of the WD40 repeat ␤-propeller structure (Couture et al. 2006; Han et al. 2006; Ruthenburg et al. 2006; Schuetz et al. 2006), which is the most common binding mode for WD40 repeat domains. Given the high sequence conservation of this protein, the surface charge of p55 is likely to be maintained across species (Supplemental Fig. 1). This charge distri- bution might be key for p55 to be in many different chromatin-modifying complexes. The electrostatic potential representation of p55 (Supplemental Fig. 2) reveals a negatively charged surface with a hole at the top and hydrophobic surface with a larger hole at the bottom of the p55. These surfaces might be used to interact with other subunits in the different complexes. Furthermore, WD40 repeat-containing proteins are known to utilize surfaces other than the one seen in this structure to recognize Figure 1. Crystal structures of p55 and a complex with histone H4 histones, possibly allowing p55 to act as a multifunc- peptide. (A) Ribbon diagram of the p55 crystal structure. p55 con- tional protein-binding platform. tains WD40 repeats forming a seven-bladed ␤-propeller structure (yellow) with an additional ␣-helix at the N terminus (blue). Disor- dered loops were connected and are shown in dashed lines. (B) Rib- The interaction between p55 and histone H4 bon diagram of the p55 bound with histone H4 peptide. Histone H4 ␤ An electrostatic potential surface representation of the (31KPAIRRLARRG41, shown in red) is bound at the side of the - propeller structure and between the N-terminal ␣-helix (blue) and histone-binding pocket of p55 shows that the binding the binding loop coming from the seventh blade of the structure. pocket consists of two distinct charged surfaces (Fig. 2a).

2 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Structure of p55

terminal ␣-helix, the side of the seventh blade, and the C-terminal one-turn ␣-helix of p55, respectively. To confirm that the interactions described above are required for histone H4 binding, we generated two p55 mutants (Fig. 3A) and examined their binding properties.

Figure 2. p55 recongizes histone H4 via the binding pocket. (a) Electrostatic surface potential representation of the binding pocket with the histone H4 peptide (shown in stick model). The binding pocket is characterized by the high negative charge on the binding loop and the hydrophobic surface on the N-terminal ␣-helix. (b) Detailed interactions between histone H4 and the p55-binding pocket. Arg39 of histone H4 is inserted in the binding pocket and interacts with the carboxylates of Asp362 and Asp365 in the binding pocket. Leu37 and Ile39 of histone H4 interact with the hydrophobic patch formed on the surface of the N-terminal ␣-helix. (c) Arg39 of histone H4 forms a hydrogen bond network with carbonyl oxgens of the backbone at the binding pocket. (d) Full-length H4 was super- imposed on the p55–H4 complex. The histone fold of histone H4 must be altered to allow p55 binding.

The surface on the binding loop from the seventh blade is highly negatively charged. The other surface formed by the N-terminal ␣-helix and the side of the seventh blade is hydrophobic. Binding of the histone H4 helix by p55 in this pocket is consistent with an in vivo study of a temperature sensitive mutant (Y41H) of Msi16, a yeast homolog of p55. This mutant showed defects in cell pro- liferation and histone deposition (Hayashi et al. 2004). Tyr41 in Msi16 corresponds to Tyr36 in p55, which is located beneath the hydrophobic patch in the binding pocket (Fig. 2b). Upon elevated temperature, this mutation might disrupt the binding pocket, suggesting that the binding pocket is functional in vivo. The two surfaces of the pocket match well with the differentially charged surfaces of the first helix of histone H4, where pairs of hydrophobic and basic amino acids (AI–RR–LA–RR) alternate making one side of the helix hydrophobic and the other side positively charged. Spe- cifically, Arg39 of histone H4 forms ionic interactions with the carboxylates of Asp362 and Asp365, and makes extensive hydrogen bonds with carbonyl oxygens from the backbone of the binding loop and the oxygen of the Gln358 side chain (Fig. 2b,c). This extensive hydrogen Figure 3. Histone-binding properties of p55. (A) Purified wild-type bond network appears to be a major contributor to his- and mutant p55. Wild-type and mutant p55 were used as analytes at tone H4 binding to p55. In contrast to these ionic inter- the concentrations of 5 nM, 10 nM, 25 nM, 50 nM, 100 nM, 250 nM, actions, the other side of the histone H4 helix makes 500 nM, 1 µM, and 2 µM on CM5 chips immobilized with full- length histone H4 (B–D); full-length mutant histone H4, H4R39A hydrophobic interactions with the binding pocket. Ile34, (E); histone H3-H4 tetramer (F–H); mutant tetramer, H3-H4R39A (I); Leu37, and Ala38 in histone H4 face a hydrophobic patch and full-length histone H3 (J–L) with 520, 430, 177, 180, and 493 RU, composed of Leu35, Phe372, and Val412 from the N- respectively.

GENES & DEVELOPMENT 3 Downloaded from genesdev.cshlp.org on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Song et al.

In one mutant, we disrupted the ionic interactions be- mutant p55 were used as analytes. While the mutant p55 tween histone H4 and the binding loop by simulta- proteins were impaired for binding to H4, these mutants neously replacing Asp362 and Arg365 with alanines. In retained similar binding ability for histone H3 as wild- another mutant, Leu35 was mutated to Ser to disrupt type p55 (Fig. 3J–L). This suggests that p55 utilizes dif- hydrophobic interactions between histone H4 and the ferent surfaces for binding to histone H3 and H4, further N-terminal ␣-helix. These residues are solvent-exposed supporting the hypothesis that p55 serves as a multi- and not involved in any stabilizing interaction within functional protein-binding platform. p55. Both mutant p55 proteins behaved identically to wild-type protein in size exclusion chromatography (data The histone H4-binding pocket is important not shown), indicating that mutating these residues did for the function of p55-containing complexes not significantly alter the folding of the protein. To examine the binding properties of the mutants, we Since p55 is a common subunit of many complexes, we used Surface Plasmon Resonance (SPR). Full-length his- examined if disrupting the binding pocket in p55 affects tone H4 was immobilized, and p55 wild-type and mutant the activity of two of these complexes. The human Hat1 proteins were used as analytes. Wild-type p55 bound full- complex is composed of the Hat1p catalytic subunit and length histone H4 tightly with a KD estimated to be in RbAp48 (the human ortholog of p55), and acetylates free the hundred nanomolar range (Fig. 3B), while we were but not nucleosomal histone H4 at Lys5 and Lys12 (Ver- unable to detect binding of mutant proteins to the im- reault et al. 1998). We compared histone acetyltransfer- mobilized histone H4, even at concentrations up to 2 µM ase (HAT) activity with Hat1 complexes containing wild- (Fig. 3C,D). This strongly suggests that the interaction type or H4-binding pocket mutants of RbAp48. Two mu- observed in the crystal structure is utilized by p55 to tants of RbAp48, equivalent to the p55 mutants, were bind to histone H4. constructed based on sequence alignment between In the reverse experiment, we examined the role of RbAp48 and p55 (Supplemental Fig. 1). Leu31 in the N- Arg39 in histone H4. The crystal structure shows Arg39 terminal helix and Asp358 and Asp361 in the binding of histone H4 inserted into the binding pocket of p55 and loop of RbAp48 were mutated to alanines. Hat1 com- forming an extensive hydrogen bond network with p55. plexes were expressed and copurified with either wild- To examine if Arg39 of histone H4 is a major contributor type or one of the mutant RbAp48 subunits (Fig. 4A). to binding, we immobilized full-length mutant (R39A) Wild-type and mutant RbAp48 form equivalently stable histone H4 on a Sensor Chip, and wild-type p55 was used complexes with the Hat1 subunit, and all complexes be- as an analyte. The R39A mutation disrupted binding, dem- haved similarly during gel filtration chromatography, in- onstrating the importance of this residue (Fig. 3E). dicating that these mutations do not disrupt folding or Previously, it was shown that p55 interacts with the complex formation. The Hat1 complexes containing mu- first helix of histone H4, which is buried in the canonical tant RbAp48 show significantly less HAT activity than nucleosome structure (Verreault et al. 1998). That find- the wild-type Hat1 complex (Fig. 4B). This supports a ing is confirmed in the crystal structure reported here, role for the H4-binding pocket of p55 in the acetyltrans- where p55 interacts with most of the surface area of the ferase activity of the Hat1 complex. first helix of histone H4. This indicates that the histone p55 binding to a buried region of histone H4 is consis- fold of H4 must be altered upon p55 binding even when tent with the function of some, but not all, p55-contain- free histone H4 is the substrate (Fig. 2d). Furthermore, ing complexes. The Hat1 and CAF-1 complexes are in- because the first helix of histone H4 is involved in volved in histone assembly; the p55 subunit in these dimerization with histone H3, p55 probably cannot bind complexes most likely binds to free histone H4. p55 is to histone H4 in canonical H3–H4 dimer (or tetramer) also found in the ATP-dependent chromatin remodeling structure, or in the nucleosome. To examine whether complexes NURF and NuRD. It is tantalizing to propose p55 can bind to the H3–H4 dimer (or tetramer), we as- that the first helix of histone H4 becomes accessible to sembled and immobilized histone H3–H4 tetramer on a p55 during the ATP-dependent chromatin remodeling Sensor Chip. Interestingly, we were able to detect binding of p55 to the immobilized H3–H4 tetramer (Fig. 3F). This suggests that the first helix of histone H4 might become accessible upon p55 binding even in a tetramer structure. Consistent with our observation, it was shown that hHat1 complex can acetylate H3–H4 tetramer in vitro (Verreault et al. 1996). One possibility, however, is that H3–H4 tetramer falls apart on the surface of the chip during the immobilization step, although it is unlikely due to the stability of histone H3–H4 tetramer. We also were able to detect residual binding of p55 mutants to histone H3–H4 tetramer, although to a significantly lesser extent than wild-type p55 (Fig. 3G,H). Considering that p55 mutants cannot bind histone H4 (see above), this binding might result from interactions with histone H3, an observation that has been made previously (Beisel et al. 2002; Wysocka et al. 2006). Consistent with this, Figure 4. HAT activities of human Hat1 complex. (A) Purified we also observed binding of wild-type p55 to the mutant hHat1 subunit alone (N-terminal His-tagged), wild-type human Hat1 complex (His tag on RbAp48), and Hat1 complexes containing tetramer H3–H4R39A (Fig. 3I). mutant RbAp48. (B) Mutations in the binding pocket of RbAp48 To examine whether p55 mutants bind histone H3, we decrease the HAT activity. (A,C) Equal amounts of human HAT immobilized full-length histone H3, and wild-type and complexes and full-length H4 were used for HAT assay.

4 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Structure of p55 process where substantial structural changes in the crystallographic refinement was done with the program CNS against the nucleosome might occur (Narlikar et al. 2002). Another native p55 data or p55–H4 data. p55-containing complex, PRC2, methylates histone H3 at Lys27 in the intact nucleosome where the first helix of Binding studies histone H4 is buried. It is therefore of interest to exam- All binding studies were performed with the SPR technique using a Bi- ine the function of p55 in this complex. acore T100 (Biacore). The surfaces were activated with an NHS and EDC mixture. Full-length histone H4, H4R39A, H3H4 tetramer, H3H4R39A To examine this, we attempted to form PRC2 com- tetramer, and full-length H3 were immobilized on a Sensor Chip CM5 plexes that contained wild-type and mutant forms of with 520, 430, 177, 180, and 493 RU, respectively. The surfaces were then p55. However, mutant forms of p55 would not stably blocked with 1 M ethanolamine solution. After immobilization, the sur- associate with the remaining subunits of PRC2 (data not face was stabilized with a regeneration buffer containing 50 mM Tris- shown). This suggests that the binding pocket of p55 HCl (pH 8.0), 1.5 M NaCl, 10 mM imidazole, and 0.05% NP40. Wild-type might be interacting with other subunits of PRC2 rather and mutant p55 (5 nM, 10 nM, 25 nM, 50 nM, 100 nM, 250 nM, 500 nM, than with histone H4 within this complex. This does not 1 µM, and 2 µM) were used as analytes for binding, and the surface was appear to be caused by a general defect in the ability of regenerated with the regeneration buffer after each injection. Biacore these mutants to form interactions, as these same mu- T100 BiaEvaluation software (GE Healthcare) was used for the data analysis. tants are able to form a stable complex with hHAT complex and bind to histone H3. This observation implies an HAT activity assay interesting function of the histone H4-binding pocket of hHat1 (N-terminal His tag) alone and Hat1 complexes were expressed in p55; it might be utilized in different ways depending Sf9 cells infected by Baculovirus and purified with Ni-affinity and size upon the role for p55 in the complex. Elucidating the exclusion chromatography. For the HAT assay, Hat1 complexes (4 pmol) functional role for p55 in PRC2 will require further bio- were incubated with tritium-labeled acetyl coenzyme A (1.25 µmol) and chemical and structural studies. free histone H4 (1 µmol) in a 20-µL final volume for 30 min at 37°C. The The results presented here reveal the molecular basis reactions were stopped with 7 µL of SDS loading buffer and loaded on for histone H4 recognition by p55. The structure of p55 SDS-PAGE gel. The gel was then transferred to Immobilon membrane bound to the first helix of histone H4 suggests that the (Millipore) and the membrane was exposed to a PhosphorImager screen. In this condition, the reaction is linear within the initial 45 min. canonical histone fold has to be altered upon p55 binding. Moreover, we have shown that the histone H4-bind- Coordinates ing pocket of p55 plays a critical role in the function of The atomic coordinates and structure factors of the p55 and the histone p55-containing complexes. Together these data suggest H4 complex will be deposited in the Protein Data Bank (accession codes that p55 might serve as a multifunctional protein inter- 3c9c and 3c99, respectively). action platform within the many p55-containing complexes. Acknowledgments Materials and methods We thank Drs. Rebecca Dunn, Ihn Sik Seong, Adam Matthews, Karim Bouazoune, and Jesse Cochrane for critical reading of the manuscript, and Protein purification Dr. Karim-Jean Armache for helping with data collection and critical Full-length Drosophila p55 with an N-terminal His tag was expressed in reading of the manuscript. We thank Dr. Bruce Stillman for providing Sf9 cells infected by Baculovirus. Infected cells were lysed and cleared by human Hat1 clone. We also thank Drs. Alexei Soares and Anand M. Saxena centrifugation. The cleared cell lysate was incubated with Ni-NTA beads for support with data collection. Data for this study were measured at (Qiagen) in the presence of 20 mM imidazole for 2 h. The Ni-NTA beads beamline X12B and X29 of the NSLS at BNL. J.J.S. is a post-doctoral were collected and washed with Buffer A containing 20 mM imidazole, fellowship recipient of the Jane Coffin Childs Memorial Fund. This work and Buffer A containing 1 M NaCl. p55 was eluted with Buffer A con- was supported by grants from the NIH (R.E.K.). taining 100 mM imidazole, and the N-terminal His tag of p55 was re- moved by TEV protease (Invitrogen). p55 was further purified with Hitrap References Q anion exchange (GE Healthcare) and Superdex 200 (GE Healthcare) size exclusion columns. p55 was concentrated up to 10 mg/mL in 50 mM Ach, R.A., Taranto, P., and Gruissem, W. 1997. A conserved family of Tris-HCl (pH 8.0) and 150 mM NaCl. Se-Met-substituted protein was WD-40 proteins binds to the retinoblastoma protein in both plants expressed in Sf9 cells in methionine-deficient media (Invitrogen) in the and animals. Plant Cell 9: 1595–1606. presence of 100 mg/L Se-methionine as described in Antipenko et al. (2003). Antipenko, A., Himanen, J.P., van Leyen, K., Nardi-Dei, V., Lesniak, J., Mutants were generated with QuickChange Site-Directed Mutagenesis Barton, W.A., Rajashankar, K.R., Lu, M., Hoemme, C., Puschel, A.W., Kit (Qiagen) and purified in the same way as the wild-type p55. et al. 2003. Structure of the semaphorin-3A receptor binding module. Neuron 39: 589–598. Crystallization and structure determination Beisel, C., Imhof, A., Greene, J., Kremmer, E., and Sauer, F. 2002. Histone Initial crystals of p55 were grown by vapor diffusion using the hanging- methylation by the Drosophila epigenetic transcriptional regulator drop method in a buffer containing 100 mM HEPES (pH 7.5), 1.4 M ammo- Ash1. Nature 419: 857–862. nium sulfate (AMS), and 3% ␤-octy-glucopyranoside. The optimal crystal- Bouveret, R., Schonrock, N., Gruissem, W., and Hennig, L. 2006. Regu- lization condition was determined as 100 mM HEPES (pH 7.5), 1.4 M AMS, lation of flowering time by Arabidopsis MSI1. Development 133: 1693– ␤ 3% -octyl-maltoside, 10 mM CdCl2, 3% ethylene glycol, and 50 mM 1702. NaCl. p55–H4 peptide complex crystals were obtained in the same con- Couture, J.F., Collazo, E., and Trievel, R.C. 2006. Molecular recognition dition as p55 alone in the presence of 2 mM H4 peptides. For cryopro- of histone H3 by the WD40 protein WDR5. Nat. Struct. Mol. Biol. 13: tection, crystals were soaked for 1 min in crystallization solution con- 698–703. taining 25% glycerol. All data were collected under cryogenic condition Czermin, B., Melfi, R., McCabe, D., Seitz, V., Imhof, A., and Pirrotta, V. (105°K) at beamline X29 or X12B at the National Synchrotron Light Source 2002. Drosophila enhancer of Zeste/ESC complexes have a histone (NSLS) at Brookhaven National Laboratory (BNL). Data were processed with H3 methyltransferase activity that marks chromosomal Polycomb HKL2000 (http://www.hkl-xray.com). Phases were calculated from the sites. Cell 111: 185–196. SAD method at the selenium peak. Initial Se sites were found using the Groth, A., Rocha, W., Verreault, A., and Almouzni, G. 2007. Chromatin program HKL2MAP (Pape and Schneider 2004) and then refined with the challenges during DNA replication and repair. Cell 128: 721–733. program Solve. Density modification was performed using the program Guitton, A.E. and Berger, F. 2005. Loss of function of MULTICOPY Resolve (Terwilliger 2000). Models were built using the program O and SUPPRESSOR OF IRA 1 produces nonviable parthenogenetic embryos

GENES & DEVELOPMENT 5 Downloaded from genesdev.cshlp.org on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Song et al.

in Arabidopsis. Curr. Biol. 15: 750–754. N.J. 2004. p55, the Drosophila ortholog of RbAp46/RbAp48, is re- Han, Z., Guo, L., Wang, H., Shen, Y., Deng, X.W., and Chai, J. 2006. quired for the repression of dE2F2/RBF-regulated genes. Mol. Cell. Structural basis for the specific recognition of methylated histone H3 Biol. 24: 9124–9136. lysine 4 by the WD-40 protein WDR5. Mol. Cell 22: 137–144. Terwilliger, T.C. 2000. Maximum-likelihood density modification. Acta Hassig, C.A., Fleischer, T.C., Billin, A.N., Schreiber, S.L., and Ayer, D.E. Crystallogr. D Biol. Crystallogr. 56: 965–972. 1997. Histone deacetylase activity is required for full transcriptional Tyler, J.K., Bulger, M., Kamakaka, R.T., Kobayashi, R., and Kadonaga, J.T. repression by mSin3A. Cell 89: 341–347. 1996. The p55 subunit of Drosophila chromatin assembly factor 1 is Hayashi, T., Fujita, Y., Iwasaki, O., Adachi, Y., Takahashi, K., and homologous to a histone deacetylase-associated protein. Mol. Cell. Yanagida, M. 2004. Mis16 and Mis18 are required for CENP-A load- Biol. 16: 6149–6159. ing and histone deacetylation at centromeres. Cell 118: 715–729. Verreault, A., Kaufman, P.D., Kobayashi, R., and Stillman, B. 1996. Hennig, L., Taranto, P., Walser, M., Schonrock, N., and Gruissem, W. Nucleosome assembly by a complex of CAF-1 and acetylated his- 2003. Arabidopsis MSI1 is required for epigenetic maintenance of tones H3/H4. Cell 87: 95–104. reproductive development. Development 130: 2555–2565. Verreault, A., Kaufman, P.D., Kobayashi, R., and Stillman, B. 1998. Huang, S., Lee, W.H., and Lee, E.Y. 1991. A cellular protein that com- Nucleosomal DNA regulates the core-histone-binding subunit of the petes with SV40 T antigen for binding to the retinoblastoma gene human Hat1 acetyltransferase. Curr. Biol. 8: 96–108. product. Nature 350: 160–162. Wade, P.A., Jones, P.L., Vermaak, D., and Wolffe, A.P. 1998. A multiple Kohler, C., Hennig, L., Bouveret, R., Gheyselinck, J., Grossniklaus, U., subunit Mi-2 histone deacetylase from Xenopus laevis cofractionates and Gruissem, W. 2003. Arabidopsis MSI1 is a component of the with an associated Snf2 superfamily ATPase. Curr. Biol. 8: 843–846. MEA/FIE Polycomb group complex and required for seed develop- Wysocka, J., Swigut, T., Milne, T.A., Dou, Y., Zhang, X., Burlingame, ment. EMBO J. 22: 4804–4814. A.L., Roeder, R.G., Brivanlou, A.H., and Allis, C.D. 2005. WDR5 Korenjak, M., Taylor-Harding, B., Binne, U.K., Satterlee, J.S., Stevaux, O., associates with histone H3 methylated at K4 and is essential for H3 Aasland, R., White-Cooper, H., Dyson, N., and Brehm, A. 2004. Na- K4 methylation and vertebrate development. Cell 121: 859–872. tive E2F/RBF complexes contain Myb-interacting proteins and re- Wysocka, J., Swigut, T., Xiao, H., Milne, T.A., Kwon, S.Y., Landry, J., press transcription of developmentally controlled E2F target genes. Kauer, M., Tackett, A.J., Chait, B.T., Badenhorst, P., et al. 2006. A Cell 119: 181–193. PHD finger of NURF couples histone H3 lysine 4 trimethylation with Kuzmichev, A., Nishioka, K., Erdjument-Bromage, H., Tempst, P., and chromatin remodelling. Nature 442: 86–90. Reinberg, D. 2002. Histone methyltransferase activity associated Zhang, Y., Iratni, R., Erdjument-Bromage, H., Tempst, P., and Reinberg, with a human multiprotein complex containing the Enhancer of D. 1997. Histone deacetylases and SAP18, a novel polypeptide, are Zeste protein. Genes & Dev. 16: 2893–2905. components of a human Sin3 complex. Cell 89: 357–364. Li, B., Carey, M., and Workman, J.L. 2007. The role of chromatin during Zhang, Y., LeRoy, G., Seelig, H.P., Lane, W.S., and Reinberg, D. 1998. The transcription. Cell 128: 707–719. dermatomyositis-specific autoantigen Mi2 is a component of a com- Lu, X. and Horvitz, H.R. 1998. lin-35 and lin-53, two genes that antago- plex containing histone deacetylase and nucleosome remodeling ac- nize a C. elegans Ras pathway, encode proteins similar to Rb and its tivities. Cell 95: 279–289. binding protein RbAp48. Cell 95: 981–991. Martinez-Balbas, M.A., Tsukiyama, T., Gdula, D., and Wu, C. 1998. Dro- sophila NURF-55, a WD repeat protein involved in histone metabolism. Proc. Natl. Acad. Sci. 95: 132–137. Muller, J., Hart, C.M., Francis, N.J., Vargas, M.L., Sengupta, A., Wild, B., Miller, E.L., O’Connor, M.B., Kingston, R.E., and Simon, J.A. 2002. Histone methyltransferase activity of a Drosophila Polycomb group repressor complex. Cell 111: 197–208. Narlikar, G.J., Fan, H.Y., and Kingston, R.E. 2002. Cooperation between complexes that regulate chromatin structure and transcription. Cell 108: 475–487. Pape, T. and Schneider, T.R. 2004. HKL2MAP: A graphical user interface for phasing with SHELX programs. J. Appl. Crystallogr. 37: 843–844. Parthun, M.R., Widom, J., and Gottschling, D.E. 1996. The major cyto- plasmic histone acetyltransferase in yeast: Links to chromatin replication and histone metabolism. Cell 87: 85–94. Qian, Y.W. and Lee, E.Y. 1995. Dual retinoblastoma-binding proteins with properties related to a negative regulator of ras in yeast. J. Biol. Chem. 270: 25507–25513. Qian, Y.W., Wang, Y.C., Hollingsworth Jr., R.E., Jones, D., Ling, N., and Lee, E.Y. 1993. A retinoblastoma-binding protein related to a negative regulator of Ras in yeast. Nature 364: 648–652. Ruthenburg, A.J., Wang, W., Graybosch, D.M., Li, H., Allis, C.D., Patel, D.J., and Verdine, G.L. 2006. Histone H3 recognition and presentation by the WDR5 module of the MLL1 complex. Nat. Struct. Mol. Biol. 13: 704–712. Schuetz, A., Allali-Hassani, A., Martin, F., Loppnau, P., Vedadi, M., Bochkarev, A., Plotnikov, A.N., Arrowsmith, C.H., and Min, J. 2006. Structural basis for molecular recognition and presentation of histone H3 by WDR5. EMBO J. 25: 4245–4252. Smith, S. and Stillman, B. 1989. Purification and characterization of CAF-I, a human cell factor required for chromatin assembly during DNA replication in vitro. Cell 58: 15–25. Sondek, J., Bohm, A., Lambright, D.G., Hamm, H.E., and Sigler, P.B. 1996. Crystal structure of a G-protein ␤␥dimer at 2.1A resolution. Nature 379: 369–374. Taunton, J., Hassig, C.A., and Schreiber, S.L. 1996. A mammalian histone deacetylase related to the yeast transcriptional regulator Rpd3p. Sci- ence 272: 408–411. Taylor-Harding, B., Binne, U.K., Korenjak, M., Brehm, A., and Dyson,

6 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Structural basis of histone H4 recognition by p55

Ji-Joon Song, Joseph D. Garlick and Robert E. Kingston

Genes Dev. published online April 28, 2008 Access the most recent version at doi:10.1101/gad.1653308

Supplemental http://genesdev.cshlp.org/content/suppl/2008/04/29/gad.1653308.DC1 Material

Related Content Diverse functions of WD40 repeat proteins in histone recognition Tamaki Suganuma, Samantha G. Pattenden and Jerry L. Workman Genes Dev. May , 2008 22: 1265-1268

Published online April 28, 2008 in advance of the full issue.

License

Email Alerting Receive free email alerts when new articles cite this article - sign up in the box at the top Service right corner of the article or click here.